Commit Graph

1058 Commits

Author SHA1 Message Date
Aleix Conchillo Flaqué
d2dfa93bf1 processors(rtvi): send bot-ready when participant joins 2024-08-16 13:58:21 -07:00
Aleix Conchillo Flaqué
fa8c6712c6 transports(daily): fix multiple DailyTransport initialization 2024-08-16 13:32:34 -07:00
Aleix Conchillo Flaqué
b57c9d569b Merge pull request #352 from pipecat-ai/aleix/rtvi-0.1
processors(rtvi): rtvi 0.1 message protocol
2024-08-15 17:35:50 -07:00
Aleix Conchillo Flaqué
f0e50ba000 Merge pull request #336 from nulyang/fix/azure-transcriptionframe
services(azure): fix TranscriptionFrame parameter type
2024-08-15 17:08:56 -07:00
Mattie Ruth
4a6638f749 Merge pull request #385 from pipecat-ai/mrkb/anthropic-beta-caching
Mrkb/anthropic beta caching
2024-08-15 18:26:51 -04:00
Aleix Conchillo Flaqué
31577252f3 processors(rtvi): handle ErrorFrames 2024-08-15 15:23:31 -07:00
Aleix Conchillo Flaqué
5d71c50080 transports(daily): make sure audio_in_task exists before canceling 2024-08-15 15:23:07 -07:00
Aleix Conchillo Flaqué
981269d594 pipeline(task): process ErrorFrame in same task and stop pipeline task 2024-08-15 15:22:40 -07:00
mattie ruth backman
848db985fc bump anthropic in 3.10 requirements 2024-08-15 16:51:48 -04:00
mattie ruth backman
d5d8e31447 add cache tokens to metrics event 2024-08-15 16:51:48 -04:00
Aleix Conchillo Flaqué
66670a2370 Merge pull request #384 from pipecat-ai/aleix/enable-prompt-caching-frames
services(anthropic): allow setting enable prompt caching via frame
2024-08-15 13:26:39 -07:00
Aleix Conchillo Flaqué
5637f349c6 services(anthropic): allow setting enable prompt caching via frame 2024-08-15 12:43:29 -07:00
Aleix Conchillo Flaqué
93248e1d00 Merge pull request #382 from pipecat-ai/khk/anthropic-beta-caching
Support for Anthropic prompt caching beta
2024-08-15 12:34:54 -07:00
Kwindla Hultman Kramer
187769357f update version number of anthropic dependency 2024-08-15 12:28:41 -07:00
Aleix Conchillo Flaqué
5be6422cc8 Revert "processors(rtvi): process options in the order they are defined"
This reverts commit 61ac83e2d9.
2024-08-15 11:51:00 -07:00
Aleix Conchillo Flaqué
8670b2d994 utils: add match_endofsentence and use it in processors 2024-08-15 11:26:25 -07:00
Aleix Conchillo Flaqué
0bc6db428d processors(rtvi): implement bot-started-speaking and bot-stopped-speaking 2024-08-15 11:05:10 -07:00
Aleix Conchillo Flaqué
67d565930e services: send TTSStartFrame/TTSStopFrame when really needed 2024-08-15 11:05:10 -07:00
Aleix Conchillo Flaqué
b2a7ff6fd3 processors(rtvi): all transport messages should be urgent 2024-08-15 11:05:10 -07:00
Aleix Conchillo Flaqué
425a730d7c transports(base_output): send urgent transport messages immediately 2024-08-15 11:05:10 -07:00
Aleix Conchillo Flaqué
84c5709722 frames: add urgent field to TransportMessageFrame 2024-08-15 11:05:10 -07:00
Kwindla Hultman Kramer
94deec01c9 okay, both files now 2024-08-15 00:57:10 -07:00
Kwindla Hultman Kramer
6e0dd4a779 Anthropic beta prompt caching 2024-08-15 00:54:43 -07:00
Kwindla Hultman Kramer
14bde340dd Merge pull request #381 from pipecat-ai/khk/anthropic-fixup-0814.2
Fixup anthropic context set_messages
2024-08-14 23:34:31 -07:00
Kwindla Hultman Kramer
253765c611 and fixing anthropic demos 2024-08-14 23:14:20 -07:00
Kwindla Hultman Kramer
2b26d7182f replaces 379 2024-08-14 22:40:09 -07:00
Aleix Conchillo Flaqué
61ac83e2d9 processors(rtvi): process options in the order they are defined 2024-08-14 22:26:49 -07:00
Aleix Conchillo Flaqué
d5c7b28cad Merge pull request #380 from pipecat-ai/aleix/rtvi-0.1-context-aggregators-updates
processors(aggregators): multiple LLM aggregators updates
2024-08-14 20:43:50 -07:00
Aleix Conchillo Flaqué
959580a708 processors(logger): fix linting 2024-08-14 20:39:24 -07:00
Aleix Conchillo Flaqué
3a5cd17ea3 processors(aggregators): multiple LLM aggregators updates 2024-08-14 20:23:18 -07:00
Kwindla Hultman Kramer
b78981bb9d Merge pull request #374 from pipecat-ai/khk/together
Together.ai service implementation with Llama 3.1 function calling
2024-08-14 17:29:07 -07:00
Kwindla Hultman Kramer
a6d90b0a00 linting fixes to anthropic.py 2024-08-14 17:27:00 -07:00
Aleix Conchillo Flaqué
67016492f2 transports(daily/helpers): add delete_room_from_url() 2024-08-14 17:14:02 -07:00
Aleix Conchillo Flaqué
2c38089527 processors(rtvi): handle incoming messages in a separate task 2024-08-14 15:34:02 -07:00
Kwindla Hultman Kramer
48f68ba6dc Service for together.ai, including Llama 3.1 function calling support 2024-08-13 15:01:54 -07:00
Aleix Conchillo Flaqué
574df4ba3d processors(rtvi): make sure to send bot-ready when transport is joined 2024-08-13 13:25:15 -07:00
Aleix Conchillo Flaqué
49ca16d125 pipeline(task): only send initial metrics frames if metrics enabled 2024-08-13 12:22:37 -07:00
Aleix Conchillo Flaqué
87525b085e processors(rtvi): linting and make send_error() public 2024-08-13 11:21:51 -07:00
Aleix Conchillo Flaqué
6b53c6add3 transports(daily): DailyTransport default DailyParams 2024-08-13 11:13:18 -07:00
Kwindla Hultman Kramer
29ca1b7855 Anthropic tool use core Pipecat pieces refactored (#369)
* processors(rtvi): rtvi 0.1 message protocol

* added a single function call handler

* wip - function calling

* fixup

* fixup

* fixup

* processors(rtvi): no need for configure_on_start()

* processors(rtvi): add new option values if they haven't been set yet

* Add the model name to the LLM usage metrics

* wip - anthropic tool calling

* still wip - anthropic tool use and vision

* anthropic tools and vision working

* anthropic tool calling and vision

* Cartesia error handling

* Anthropic tool use core Pipecat pieces refactored as per plan

* aleix has good ideas

* Usage metrics for Anthropic LLMs

* fix function call result state not getting cleared bug

* Pass **kwargs through from AnthropicLLMService constructor

* about to tinker with anthropic

* added openai function calling

* openai function calling

* fixup

---------

Co-authored-by: Aleix Conchillo Flaqué <aleix@daily.co>
Co-authored-by: Chad Bailey <chadbailey@gmail.com>
Co-authored-by: mattie ruth backman <mattieruth@gmail.com>
Co-authored-by: chadbailey59 <chadbailey59@users.noreply.github.com>
2024-08-13 13:01:24 -05:00
Aleix Conchillo Flaqué
a42d0c9907 processors(rtvi): add interrupt_bot() 2024-08-13 09:22:43 -07:00
marcus-daily
8bc6ceaa3d Fixing pep8 2024-08-13 15:32:23 +01:00
marcus-daily
0b8a1ab5d1 Handle describe-actions message 2024-08-13 15:32:23 +01:00
Brian Hill
358c287db2 chore: Enable build without git 2024-08-12 11:38:41 -04:00
Brian Hill
2e68453655 Merge pull request #371 from pipecat-ai/cbrianhill/allow-build-without-git
chore: Enable build without git
2024-08-12 10:15:55 -04:00
Brian Hill
89b8a9de7d chore: Enable build without git 2024-08-12 09:36:25 -04:00
Aleix Conchillo Flaqué
c4c2058df9 processors(rtvi): handle frames pushed from outside in order 2024-08-11 23:09:11 -07:00
Aleix Conchillo Flaqué
0d85c0085f processors(rtvi): interrupt the bot if a new config is received 2024-08-11 23:09:11 -07:00
Mattie Ruth
6fa8a8f84f Merge pull request #365 from pipecat-ai/ruthless/metrics 2024-08-11 20:35:05 -04:00
mattie ruth backman
a97775bff3 Add the model name to the LLM usage metrics 2024-08-11 12:08:46 -04:00